1

HARVARD LAW SCHOOL

CAMBRIDGE · MASSACHUSETTS · 02138 October 25, 2016 To the Honorable City Council Chair Amy Foster and the Honorable City Council, I am writing to express my support for the city’s proposed ordinance regarding political spending by foreign-influenced corporations as a critical tool for uncovering foreign influences prohibited under federal law, as well as to explain how corporations can obtain responsive information about the foreign national status of shareholders, as mandated by the ordinance’s certification requirement. Background I am the John F. Cogan, Jr. Professor of Law and Economics at Harvard Law School, and the Research Director of the Center on the Legal Profession. Before joining Harvard, I was a partner at Wachtell, Lipton, Rosen & Katz, specializing in financial institutions and mergers and acquisitions. I have testified before the United States Congress, am on the Investor Advisory Committee of the Securities and Exchange Commission (SEC), and have provided consulting services to the U.S. Department of Justice, the U.S. Department of Treasury, the New York Stock Exchange, and participants in the financial markets, including individuals, mutual funds, hedge funds, investment banks, commercial banks, and private equity funds. In June 2016, I testified by invitation at a forum on “Corporate Political Spending and Foreign Influence” at the Federal Election Commission. Foreign corporate spending in American elections Since the Supreme Court’s 2010 Citizens United decision invalidated restrictions on corporate political spending,1 the possibility that American elections could be influenced by foreign interests via corporations has attracted considerable public and policymaker interest. Foreign governments, foreign-based companies, and people who are neither U.S. citizens or permanent residents are

1 Citizens United v. FEC, 558 U.S. 310 (2010).

2

currently barred by federal law from contributing or spending money in connection with federal, state, or local elections.2 Unfortunately, Citizens United created a loophole to this ban: these foreign entities can invest money through U.S.-based corporations that can – as a result of the decision – then spend unlimited amounts of money in American elections. The policy interest in regulating foreign influence need not rest on the idea that foreign investors are tied to hostile governments that are actively trying to undermine the democracy or economy of the United States. Rather, it may also or separately rest on the observation that foreign nationals (even those in countries that are staunch U.S. allies) are not part of the U.S. polity. Democratic self-governance presumes a coherent and defined population to engage in that activity. Foreign nationals have a different set of private interests than their U.S. counterparts, as regards a range of policies, such as defense, environmental regulation, and infrastructure. Few dispute the idea that a given government may properly seek to limit foreign influence over, in the words of the U.S. Supreme Court, “activities ‘intimately related to the process of democratic self-government.’”3 There is nothing particularly surprising or pernicious about this fact. Foreign and domestic interests predictably diverge. Depending on the degree of their influence, foreign governments (or their agents, such as sovereign wealth funds), foreign corporations, or other foreign investors might be able to leverage ownership stakes in U.S. corporations to affect corporate governance. Through that channel, they could influence corporate political activity in a manner inconsistent with democratic self-government, or at least out of alignment with the interests of U.S. voters. Every country regulates some types of foreign and domestic business activities differently. In many domains of the American economy, long-standing statutes, regulations, and legal traditions treat foreign companies or foreign-influenced companies differently than domestic companies. The United States has specific foreign restrictions across a number of different industries. In shipping, aircraft, telecom, and financial services, laws governing all of these industries limit or regulate foreign ownership or control. Some ban foreign ownership completely, and, for some, foreign ownership or control triggers special government approval procedures. 2 52 U.S.C. § 30121(a). This prohibition was upheld by a unanimous U.S. Supreme Court in 2012. See Bluman v. FEC, 132 S. Ct. 1087 (2012). 3 Bluman v. FEC, 800 F. Supp. 2d 281, 287 (D.D.C. 2011) (three-judge court) (quoting Bernal v. Fainter, 467 U.S. 216, 220 (1984)), aff’d, 132 S.Ct. 1087 (2012).

3

The same spirit of those bodies of law should inform regulation of election spending by foreign-influenced corporations. Since Citizens United opened the door for political activity by corporations, some corporations of which ownership or control is likely held in significant part by foreign entities have devoted considerable financial resources to influencing American elections. In practice, the policy preferences of foreign-influenced corporations are sometimes clear from public sources. This past May, Uber and Lyft spent over $9 million on a ballot initiative in Austin, Texas that would have overturned an ordinance passed by the Austin City Council requiring the companies’ drivers to submit to fingerprint-based criminal background checks.4 Weeks later, Uber disclosed that the Saudi Arabian government had invested $3.5 billion in the company, giving the Kingdom over five percent ownership and a seat on its board of directors.5 Last month, the multinational “homestay” corporation Airbnb responded to the New York Legislature’s growing interest in regulating the industry by arming a super PAC with $11 million to influence New York’s legislative races.6 Airbnb – a privately held company – is partly owned by Moscow-based DST Global.7 More strikingly, an investigation by The Intercept uncovered how APIC, a San Francisco-based company described as “controlled,” “owned,” and even “100 4 Nolan Hicks, “Prop 1 campaign crosses $9 million threshold,” AUSTIN-AMERICAN STATESMAN, May 9, 2016, http://atxne.ws/29pbFBk. 5 See Elliot Hannon, “Saudi Arabia Makes Record $3.5 Billion Investment in Uber,” SLATE, June 1, 2016, http://slate.me/1UvvM3x. Uber also spent roughly $600,000 on a 2015 voter referendum in Seattle. See Karen Weise, “This is How Uber Takes Over a City,” BLOOMBERG, June 23, 2015, http://bloom.bg/1Ln2MaN. 6 Kenneth Lovett, “Airbnb to spend $10 on Super PAC to fund pre-Election day ads,” N.Y. DAILY NEWS, Oct. 11, 2016, http://www.nydailynews.com/news/politics/airbnb-spend-10m-super-pac-fund-pre-election-day-ads-article-1.2825469. 7 See Dan Primack, “Yuri Milner adds $1.7 billion to his VC war chest,” FORTUNE, Aug. 3, 2015, http://fortune.com/2015/08/03/yuri-milner-adds-1-7-billion-to-his-vc-warchest/ (DST Global is Moscow based); Scott Austin, “Airbnb: From Y Combinator to $112M Funding in Three Years, The Wall Street Journal, July 25, 2011, http://blogs.wsj.com/venturecapital/2011/07/25/airbnb-from-y-combinator-to-112m-funding-in-three-years/ (DST Global is a major investor in Airbnb).

4

percent owned” by Gordon Tang and Huaidan Chen—two Chinese citizens with permanent residence in Singapore—gave $1.3 million to a super PAC that had supported Jeb Bush’s run for president.8 Though the story made headlines, it echoes similar, yet less publicized, efforts to influence high-profile state and national races. For example, in 2012, a Connecticut-based subsidiary of a Canadian insurance and investment corporation gave $1 million to the pro-Mitt Romney super PAC Restore Our Future.9 In 2013, a New Jersey-based subsidiary of a Chinese-owned business contributed $120,000 directly to Terry McAuliffe’s gubernatorial campaign in Virginia.10 Ballot initiatives have been particularly strong magnets for spending by multinational corporations. This past August, American Electric Power, Limited Brands, and Nationwide Insurance spent a combined $275,000 against a municipal initiative aimed at reconfiguring the Columbus City Council.11 In 2012, a Los Angeles County ballot measure, the “Safer Sex in the Adult Film Industry Act,” attracted over $325,000 from two companies tied to a Luxembourg corporation that ran adult webpages.12 The company’s then-CEO was a German national.13 That same year, a statewide ballot initiative in California that would have required all foods containing genetically modified organisms to be labeled as such attracted $45 million in spending by multinationals such as Monsanto and DuPont.14 Opponents of the measure

8 Jon Schwartz & Lee Fang, “The Citizens United Playbook,” THE INTERCEPT, Aug. 3, 2016, http://bit.ly/2auW75p. 9 Michael Beckel, “Foreign-Owned Firm Gives $1 Million to Romney Super-PAC,” MOTHER JONES, Oct. 5, 2012, http://www.motherjones.com/politics/2012/10/canadian-foreign-donation-super-pac-restore-our-future. 10 John Schwartz, “Va. Gov. Terry McAuliffe Took $120K from a Chinese Billionaire—but the Crime Is That It Was Legal,” THE INTERCEPT, June 1, 2016, http://bit.ly/1XPvuXN. 11 Lucas Sullivan, “Follow the money flowing to ward initiative campaigns in Columbus,” THE COLUMBUS DISPATCH, July 22, 2016, http://bit.ly/2ahlSpq. 12 See Ciara Torres-Spelliscy, “How a Foreign Pornographer Tried to Win a U.S. Election,” THE BRENNAN CENTER FOR JUSTICE, Nov. 6, 2015, http://bit.ly/29pesu2. 13 Id. 14 Suzanne Goldenberg, “Prop 37: food companies spend $45m to defeat California GM label bill,” THE GUARDIAN, Nov. 5, 2012, http://bit.ly/29I3SE7.

5

spent five times more than its supporters, and ultimately defeated it by a 53-47 margin.15 Of course, not all politically active corporations are owned or controlled in significant part by foreign entities. Many privately held companies are owned directly by one or a small number of U.S. citizens. Among U.S. public companies, foreign ownership varies. I have recently completed research (attached as an appendix to this letter) finding that, among publicly traded corporations in the Standard & Poor’s (S&P) 500 index, only one in eleven (~9 percent) has a foreign institutional investor with more than five percent of the company’s voting shares. (Five percent is the threshold at which federal securities law requires public disclosure of large stockholdings of US public companies.16) Other corporations may have foreign ownership at substantial levels (e.g., 20 percent or more) that would make unaffiliated foreign investors potentially capable of exerting influence on the corporate political spending, but lack any single foreign owner that holds at least five percent of total stock. Regulating foreign corporate spending Cities can simultaneously welcome foreign investment without exposing themselves to the risk of foreign money influencing their elections. The proposed ordinance addresses this issue through a requirement that a corporation spending money in city elections certify that it is not a “foreign-influenced corporation” – a definition based, in part, on the extent of foreign ownership of corporate stock. The proposed ordinance is a reasonable response to an increasingly localized problem, and is constitutional under the Court’s decision in Citizens United. The remainder of this letter details how this certification requirement could operate.

15 Id. 16 Under Section 13(d) of the Securities Exchange Act of 1934 (as amended by the Williams Act), any person or group of persons who acquire beneficial ownership of more than five percent of the voting class of the equity of a corporation that is listed or otherwise required to register as a “public” company under that law, must, within ten days, report that acquisition to the Securities and Exchange Commission (SEC) via Schedule 13D (or, in some cases, Schedule 13G). See 15 U.S.C. § 78m(d); 17 C.F.R. §§ 240.13d-1, 240.13d-101.

6

The mechanics of the ordinance’s certification requirement 1. Ownership of corporate stock

To begin, as a general matter, corporate stock may be “owned” in three different forms. First, many companies that have one or a relatively small number of shareholders hold paper stock certificates. Among larger, stock exchange listed companies, with numerous owners, such direct ownership is rare, and increasingly so. At such companies, shares are more commonly held in “street name” through a broker (e.g., Fidelity or Charles Schwab). In these instances, the name on the stock certificate is actually the broker, but the broker keeps track in a database of how many shares belong to each client. Clients who hold shares in street name are “beneficial owners” under SEC rules, can direct brokers how to vote or sell shares, and can participate in corporate governance. Most shares of large, listed companies, however, are now held by separate legal entities, such as mutual funds, pension funds, insurance companies, and hedge funds. As an economic matter, these entities hold stock on behalf of their clients or beneficiaries. However, as a legal matter, the investment entities themselves are the owners of the stock, and they do not pass through to beneficiaries either the right to vote or the right to sell the shares of the stock that the entity purchases. Individuals whose wealth is invested through these types of institutional investments cannot exercise voting rights associated with the shares. Instead, those rights are exercised by the management of the institutions.

2. Determining shareholders Most corporate stock is not traded on public markets. As of 2012, more than five million corporations filed U.S. income tax returns. Only about 4,000 corporations were listed on a U.S. stock exchange – less than 0.1 percent of corporations that filed tax returns. Of the rest, many are owned by a single shareholder, or are beneficially owned by up to 500 individual owners. (SEC rules generally require public registration and disclosure for companies with more than 500 owners and $10 million in assets.) Companies without public markets are still large and have substantial numbers of shareholders. Examples include Cargill, with revenues exceeding $130 billion and over 200 shareholders, and Mars, with revenues exceeding $33 billion and over 45 shareholders. Because shares of such companies do not trade freely in the public markets, such companies generally can and do track the identity of their shareholders directly. For corporations listed on public markets, shares trade in significant volume—

7

thousands of shares per day. However, publicly traded corporations have the ability to ascertain the exact ownership of their shares as of any arbitrary “record date.” In fact, this happens at least annually, because companies are required by corporate law to have annual shareholder meetings, for which they must set a record date to determine which shareholders are eligible to attend and vote at the meeting. In fact, record dates are set and shareholder lists are created more frequently than that at many public companies, to allow for votes on off-cycle events, such as a merger proposal or charter amendments, which are brought to a vote at special meetings. Consequently, the ability to determine record stock ownership as of a given date is essential to the basic governance of corporations.

Few if any publicly traded corporations engage in the process of determining their record shareholders for a given record date themselves. They use an intermediary – most commonly, American Stock Transfer (AST) – that is dedicated to this function. Under state law, shareholders seeking to file a derivative suit or solicit shareholder support for a shareholder resolution or proxy contest can also obtain the list of shares using the same method. A corporation that needs the list of shareholders as of a specific date would engage AST to produce the list of shareholders as of that date. Under SEC rules, public companies also reach out beyond their record holders to the beneficial owners of broker- or bank-owned stock, and engage AST to contact banks, brokers or other intermediaries that are nominally record owners. Those firms, in turn, provide information about non-objecting beneficial owners to AST, which then compiles it and provides it to the corporation. Typically, banks, brokers and other intermediaries provide AST (and the corporation) with non-objecting client names, addresses, shares held, and purchase dates (which could be multiple blocks if a given shareholder bought multiple blocks of shares over time). In addition to these basic corporate and securities law mechanisms, Section 13 of the federal Securities Exchange Act of 1934 requires any person or group of persons who acquire beneficial ownership of more than five percent of the voting class of a listed corporation’s equity to within ten days report that acquisition to the SEC on a Schedule 13D (or, in some cases, Schedule 13G).17 These acquisitions are, in turn, made public by the SEC, and available through the SEC’s EDGAR online database.

17 See 15 U.S.C. § 78m(d); 17 C.F.R. §§ 240.13d-1, 240.13d-101.

8

3. Determining whether shareholders are “foreign nationals” As just described above, acquisitions of five percent or more of the stock of public U.S. companies must already be disclosed under SEC rules, including the identity of the purchaser’s citizenship.18 Thus, the information is already publicly available for five percent blockholders of public companies. Outside of the blockholder context, for most purposes, corporations typically do not inquire into the citizenship or permanent residency status of shareholders. Many brokerage firms impose restrictions on non-citizens, or specifically limit their customers to citizens or permanent residents. A 2012 sampling of major brokers by financial markets reporter Matt Krantz found divergence in practices:

For instance, at Fidelity, the company says only U.S. citizens may open an account. . . . Over at TD Ameritrade, investors do not need to be a U.S. citizen to open an account. With that said, the stipulations and requirements vary dramatically based on the country the resident lives in and the potential customers’ nationality, the company says. . . . Similarly at E-Trade, the brokerage has different rules based on the country. . . . The rules vary widely based on the nationality of the person wanting the account . . . . TradeKing requires investors, including U.S. citizens, to be U.S. residents to establish the account. It makes an exception for customers who are living abroad and have a valid U.S. military or government address. Investors who are not U.S. citizens, yet reside legally in the U.S., may open an account if they have a Social Security number and aren’t from 27 specific [prohibited] countries . . . .19

The process of ascertaining the foreign national status of shareholders could be simple in many cases. If a publicly traded corporation asks American Stock Transfer to produce its list of shareholders (or just those shareholders who are foreign nationals), and AST in turn asks Fidelity, Fidelity’s citizens-only customer policy would enable it to truthfully and simply answer that zero percent of the company’s shares held through Fidelity are held by foreign nationals.

18 See 17 C.F.R. § 240.13d-101 (item #6, requiring reporting of “Citizenship or place of organization”). 19 Matt Krantz, USA TODAY, “U.S. online brokerage options are limited for foreigners,” http://usat.ly/KXpDan (May 16, 2012).

9

Similarly, where stock is held by a non-human shareholder, such as another corporation, the “foreign national” status of that corporation can be ascertained readily by examining its place of incorporation and principal place of business. The proposed ordinance counts stock owned by domestic subsidiaries of foreign parent corporations the same as stock owned by foreign corporations. (In the terms of the ordinance, either would be defined as a “foreign owner.”) To the extent that a U.S. subsidiary of a foreign corporation has the potential to influence U.S. portfolio companies in which it invests, it has the potential to do so at the foreign parent’s bidding or with the foreign parent’s approval. However, the ordinance does not require piercing through the beneficial ownership of institutional entities such as mutual funds. For the ordinance’s purpose, corporate stock owned by a mutual fund is not corporate stock held by a foreign national, even if many of the mutual fund’s customers are themselves foreign nationals, as long as the advisor to the fund is a U.S. entity (a fact that can be readily determined with public information). This is a reasonable approach, because customers of mutual funds cannot themselves directly participate in governance of the corporation actually spending money in a city election. Instead, it is the management of the advisory firm that plays that role.

4. “Due inquiry” Importantly, the ordinance addresses any remaining possible difficulties that U.S. corporations might have in certifying as to whether they are foreign-influenced. As noted above, some brokerage firms allow foreign nationals to buy stock of U.S. companies through them, and they may not report citizenship information about such customers to the corporations in which they invest. Thus, it may not be possible for every corporation to verify the U.S. or foreign national status of all of its shareholders with complete confidence. (Note, however, that the ordinance does not actually require a corporation to verify all of its shareholders’ statuses: Given the 20 percent, “aggregate” threshold, verifying that just over 80 percent of shareholders are not foreign owners would be sufficient.) However, given this possibility, it is reasonable for the proposed ordinance to impose a certification requirement that specifies that the chief executive officer of the corporation certify that the information is provided after “due inquiry.”

10

The “due inquiry” standard is familiar from securities law,20 as well as from other areas of law with which corporate executives are acquainted.21 It imposes only the customary obligation to make such reasonable inquiry as the corporation would do in any event. Thus, the ordinance does not impose a meaningful additional information-gathering cost beyond what it would already be required to do under existing law. Conclusion The ordinance is a reasonable solution to the risk of foreign influence in local elections through corporate political spending. The ordinance is constitutional under Citizens United, and reasonable from a corporate and securities law perspective. The ordinance would only apply to corporations that spend $5,000 or more on city elections. The ordinance imposes no obligations on corporations that spend less than that (or nothing at all) on city elections. For those corporations that do spend more than that, the certification required by the proposed ordinance is practicable and reasonable for both privately and publicly traded corporations, conditioned as it is on corporations engaging in “due inquiry,” a standard that will not add material costs to the information-gathering and record-keeping corporations already engage. If you have any further questions, please let me know. Sincerely,

John F. Cogan, Jr. Professor of Law and Economics Harvard Law School 20 See, e.g., 17 C.F.R. § 275.206(4)-2(a)(3). 21 See, e.g., SRI Int’l, Inc. v. Advanced Tech. Labs., Inc., 127 F.3d 1462, 1464–65 (Fed. Cir. 1997) (in patent law, standard for whether infringement was “willful” is “whether the infringer, acting in good faith and upon due inquiry, had sound reason to believe that it had the right to act in the manner that was found to be infringing”); Black Diamond Sportswear, Inc. v. Black Diamond Equip., Ltd., No. 06-3508-CV, 2007 WL 2914452, at *3 (2d Cir. Oct. 5, 2007) (“A trademark owner is “‘chargeable with such knowledge as he might have obtained upon [due] inquiry.’”) (quoting Polaroid Corp. v. Polarad Electronics Corp., 182 F. Supp. 350, 355 (E.D.N.Y. 1960)) (alteration in original).

Quantifying foreign institutional block ownership

at publicly traded U.S. corporations

John C. Coates IV, Ronald A. Fein, Kevin Crenny, and L. Vivian Dong

Free Speech For People Issue Report 2016-01

October 2016 !

Quantifying foreign institutional block ownership at publicly traded U.S. corporations John C. Coates IV Ronald A. Fein Kevin Crenny L. Vivian Dong Harvard Law School Program on Corporate Governance Discussion Paper No. 2016-[XXXX] Free Speech For People Issue Report No. 2016-01

1

QUANTIFYING FOREIGN INSTITUTIONAL BLOCK OWNERSHIP AT PUBLICLY TRADED U.S. CORPORATIONS

John C. Coates IV,1 Ronald A. Fein,2 Kevin Crenny,3 and L. Vivian Dong4

Abstract

This short technical report provides an empirical analysis of the level of foreign institutional block ownership at a broad set of publicly traded corporations. Disclosed institutional blockholders of every company in the Standard & Poor’s 500 index are analyzed to determine if these blockholders were foreign entities or were majority owned or controlled by foreign entities. Roughly one in eleven (9%) companies in the S&P 500 has one or more foreign institutions each owning five percent or more blocks of stock, nine have foreign institutions with ten percent or more blocks, five have a foreign institution with more than fifteen percent, and three have foreign institutions with more than 20% blocks. Three firms have multiple foreign institutional blockholders. The descriptive data reported here may assist lawmakers, analysts, and investors in assessing the effects of globalization of capital markets and the interaction of country and governance risk, and in developing policies. Among other things, these data may inform debates on the degree to which domestic political spending by U.S. corporations conveys any potential for foreign influence through governance, and the likely costs and benefits of disclosure laws regarding such influence.

Introduction

Since the Supreme Court’s 2010 Citizens United decision invalidated restrictions on corporate political spending,5 considerable public and

1 John F. Cogan Professor of Law and Economics, Harvard Law School. Professor Coates serves on Free

Speech For People’s unpaid Legal Advisory Committee. For disclosure of financial interests potentially relevant to this paper, see hls.harvard.edu/faculty/directory/10170/Coates.

2 Legal Director, Free Speech For People. For more information on Free Speech For People, a public interest advocacy organization formed on the day of the Citizens United decision, see http://www.freespeechforpeople.org. Free Speech For People supports legislation to regulate political spending by corporations with significant foreign ownership, including the local measure described in note 10.

3 J.D. candidate, Harvard Law School, 2018. 4 J.D. candidate, Harvard Law School, 2018. 5 Citizens United v. FEC, 558 U.S. 310 (2010).

2

policymaker interest has developed in the potential for U.S. elections to be influenced by foreign interests through U.S. corporations.6 On the one hand, existing federal law (the Federal Election Campaign Act) already prohibits political spending in federal, state, or local elections by corporations that are incorporated outside the U.S., or which have their principal place of business abroad.7 On the other hand, current law still allows substantial avenues for foreign influence over corporate political spending by U.S.-incorporated and -based corporations.8 Such influence could arise from board representation, manager control, contracts (including lending arrangements), or ownership of significant blocks of stock. This paper focuses on the last potential channel for influence. Lawmakers in Congress and members of the Federal Election Commission have expressed interest in addressing this phenomenon. As of yet federal reform proposals have failed to advance.9 A more likely near-term prospect for new policy measures is at the state and local level. Local governments are now contemplating measures to address this concern.10

6 Compare Citizens United, 558 U.S. at 362 (“We need not reach the question whether the Government has a compelling interest in preventing foreign individuals or associations from influencing our Nation's political process.”) with id. at 424 (Stevens, J., dissenting) (“If taken seriously, our colleagues' assumption that the identity of a speaker has no relevance to the Government's ability to regulate political speech . . . would appear to afford the same protection to multinational corporations controlled by foreigners as to individual Americans. . . .”); see also President Barack Obama, Remarks by the President in State of the Union Address (Jan. 27, 2010), https://www.whitehouse.gov/the-press-office/remarks-president-state-union-address (“With all due deference to separation of powers, last week the Supreme Court reversed a century of law that I believe will open the floodgates for special interests –- including foreign corporations –- to spend without limit in our elections.”).

7 52 U.S.C. § 30121; see also Bluman v. FEC, 800 F. Supp. 2d 281 (D.D.C. 2011) (three-judge court) (upholding prohibition on individual foreign nationals making independent expenditures against First Amendment challenge), aff’d mem., 132 S. Ct. 1087 (2012).

8 See Jon Schwarz & Lee Fang, Cracks in the Dam: Three Paths Citizens United Created for Foreign Money to Pour Into U.S. Elections, The Intercept (Aug. 3, 2016), http://interc.pt/2aA8QP8; see also FEC, Advisory Opinion No. 2006-15 (TransCanada) (May 19, 2006), http://saos.fec.gov/aodocs/2006-15.pdf (providing pre-Citizens United guidance for internal corporate decision-making processes regarding state and local political contributions by domestic subsidiary of foreign corporation).

9 See DISCLOSE Act, H.R. 5175 (111th Cong.), https://www.congress.gov/bill/111th-congress/house-bill/5175; Commissioner Ellen L. Weintraub, FEC, Proposal to launch rulemaking to ensure that U.S. political spending is free from foreign influence (Sept. 9, 2016), http://go.usa.gov/xkpFX; Commissioner Ann M. Ravel, FEC, Proposal to Rescind Advisory Opinion 2006-15 (TransCanada) (Aug. 9, 2016), http://go.usa.gov/xkpFQ; FEC, Forum: Corporate Political Spending and Foreign Influence (June 23, 2016), http://go.usa.gov/xkpFP.

10 See Lisa Wheeler-Brown & Karen Lieberman, St. Pete, let’s take a bite out of Citizens United, The Weekly Challenger (Sept. 16, 2016), http://theweeklychallenger.com/st-pete-lets-take-a-bite-out-of-citizens-united/ (describing proposed ordinance in St. Petersburg, Florida that would, inter alia, regulate political spending in city elections by corporations with significant foreign ownership); Darden Rice & Scott Greytak, Keep super PAC cash out of St. Petersburg elections, Tampa Bay Times (July 15, 2016), http://www.tampabay.com/opinion/columns/column-keep-super-pac-cash-out-of-st-petersburg-elections/2285582 (similar).

3

The policy interest in foreign ownership need not rest on the idea that foreign investors are tied to hostile governments that are actively trying to undermine the democracy or economy of the United States. Rather, it may also or separately rest on the observation that foreign nationals (even those in countries that are staunch U.S. allies) are not part the U.S. polity. Democratic self-governance presumes a coherent and defined population to engage in that activity. Foreign nationals have a different set of private interests than their U.S. counterparts, as regards a range of policies, such as defense, environmental regulation, and infrastructure. Few dispute the idea that a given government may properly seek to limit foreign influence over “activities ‘intimately related to the process of democratic self-government.’”11 Depending on the degree of their influence, foreign governments (or their agents, such as sovereign wealth funds), foreign corporations, or other foreign investors might be able to leverage ownership stakes in U.S. corporations to affect corporate governance. Through that channel, they could influence corporate political activity in a manner inconsistent with democratic self-government, or at least out of alignment with the interests of U.S. voters. It would also be reasonable to at least consider regulations that make U.S. voters aware, through disclosure, of which U.S. companies were subject to significant foreign influence. Such disclosures would be useful not simply as a matter of corporate governance policy (as under current SEC regulations), but in television ads or other specific communications in the electoral arena. The prospect of such concerns about political influence should also be of interest to investment researchers and analysts. Actual or perceived foreign influence can create a political demand for or expectation of regulation, and with it the risk of political backlash. For example, the fact that foreign banks benefited from the bailouts in the financial crisis stimulated some degree of public criticism, and may have contributed to political demand for regulatory policies limiting discretion by government agencies to engage in future market interventions. Beyond policy and politics, a better understanding of the ways in which capital integration and the growth of global institutional investors contribute to cross-border influence over portfolio companies is of general interest to economists and business scholars. This paper addresses one aspect of the policy questions raised by the potential for foreign influence over U.S. elections: quantifying foreign institutional ownership of voting shares of U.S.-based corporations at levels sufficient to have a meaningful degree of influence in corporate governance of large and well-resourced public companies — i.e., foreign blockholders. Remarkably, we are unaware of any recent empirical analysis of a broad set of corporations to ascertain the frequency and level of foreign block ownership. We examined all

11 Bluman, 800 F. Supp. 2d at 287 (quoting Bernal v. Fainter, 467 U.S. 216, 220 (1984)).

4

505 companies currently in the Standard & Poor’s (S&P) 500 Index to assess whether they had blockholders that were themselves foreign entities or were majority owned or controlled by foreign entities. While one motivation for this paper relates to corporate political spending, the generality of the empirical research may also be useful to other areas of law (e.g., telecommunications) or corporate governance, where significant foreign blockholder ownership of U.S.-based corporations may be relevant. A stylized and largely uncontested fact is that institutional shareholders — the most likely to be blockholders of U.S. public companies — are increasingly influential in the governance of those companies.12 Various changes in markets and regulation have increased the ability of such institutions to encourage, pressure or force boards to adopt policies and positions that twenty years ago would have been beyond their reach. Board members are spending increased amounts of time responding to and directly “engaging” with blockholders. While in the past legal regimes tested “control” of foreign nationals at higher levels of ownership – majority voting power, or 25% blocks for example – those regimes may no longer catch the new forms of institutional influence. Finally, the descriptive data reported here may also assist researchers, analysts and investors in assessing the effects of globalization of capital markets and the interaction of country and governance risk. Most scholarship on globalization has focused on fundamental business integration, trade flows, and their impacts on employment and production. To the extent it has been studied at all, capital market integration has been analyzed mostly in the aggregate. It has not focused on the effect of globalized capital flows on corporate governance, or on blockholders, boards, and managers as separate units of observation and interest. To the extent that corporate governance matters to real economic decisions, the fact, causes, and impacts of globalization of blockholders should be a research subfield in its own right. The paper proceeds as follows. In Part I, we briefly describe our methods. In Part II, we present our findings. We conclude with a brief discussion of implications and future research questions.

12 See John C. Coates IV, Thirty Years of Evolution in the Roles of Institutional Investors in Corporate

Governance, in Research Handbook on Shareholder Power 79-99 (Edward Elgar Publishing, eds., J. G. Hill and R.S. Thomas 2015).

5

I. Methodology

A. Defining foreign investors We developed a two-part definition of a foreign investor for purposes of this analysis. First, we count any investor that would meet the federal definition of a “foreign national” used for purposes of the Federal Election Campaign Act (FECA) — viz., a foreign government, a corporation incorporated or having its principal place of business in a foreign country, or an individual who is neither a U.S. citizen nor a lawful permanent resident.13 Second, we also count any investor that is not itself a foreign national under that definition, but which is majority owned or controlled by a foreign national. Our definition is purposefully broader than the statutory definition, because majority owners or controllers can clearly exert the same kind of influence as a corporation can exert directly. That is because a foreign national with such a level of ownership or control can practically direct the subsidiary investor’s governance activities with respect to the ultimate corporation being analyzed. (In fact, one could argue for broadening this definition even further than we did, in that shareholders with less than majority stakes can effectively control portfolio companies, as well. Our data should thus be understood as putting a lower bound on potential foreign influence through ownership.) This second part of the definition is relevant because, in many cases, foreign investment in U.S. securities occurs through subsidiaries. For example, consider Massachusetts Financial Services Company (recently renamed MFS Investment Management), a major investment management firm with substantial blockholding positions in many U.S. equities. MFS was founded in 1924 in Massachusetts, maintains its headquarters in Boston, and in many respects is as American as apple pie. It would not itself qualify as a “foreign national” under FECA. However, in 1982 it was acquired by Sun Life Financial, Inc., a Canadian firm headquartered in Toronto.14 However similar to the U.S., Canada is another country. To the extent that MFS has the potential to influence portfolio companies in which it invests, it has the potential to do so at Sun Life’s bidding or with Sun Life’s approval. (To be clear, we present no direct evidence that any foreign company has used such a potential for influence; rather, we note the possibility and estimate the frequency of that potential.) Consequently, our analysis treats MFS as functionally equivalent to Sun Life Financial, and therefore a foreign blockholder.

13 See 52 U.S.C. § 30121(b); 22 U.S.C. § 611(b). 14 See Wikipedia, MFS Investment Management,

https://en.wikipedia.org/wiki/MFS_Investment_Management (last visited Oct. 18, 2016).

6

B. Defining blockholder threshold

We then selected a five percent threshold definition for blockholders. While there is no magic in a five percent threshold, it is a widely used standard in corporate governance literature and in corporate and securities law and regulation.15 With five percent of shares, an investor is in a position to affect corporate governance through both formal (e.g., board elections, bylaw or charter amendments, fiduciary duty or other lawsuits) and informal (e.g., exit or threat of exit, withhold vote campaigns, and non-binding shareholder resolutions) mechanisms.16 Five percent is the threshold at which federal securities law requires disclosure to the Securities Exchange Commission. Under Section 13(d) of the Securities Exchange Act of 1934 (as amended by the Williams Act), any person or group of persons that acquires beneficial ownership of more than five percent of the voting class of the equity of a corporation that is listed or otherwise required to register as a “public” company under that law, must, within ten days, report that acquisition to the SEC via Schedule 13D (or, in some cases, Schedule 13G).17 The five percent threshold in the Williams Act has both conceptual and pragmatic advantages for present purposes. Conceptually, it validates selection of the threshold as a non-arbitrary value drawn from longstanding federal securities law. Pragmatically, it means that five percent blockholders can (presuming compliance) be readily ascertained from publicly available sources that draw their data from legally mandated SEC filings.18 For

15 See, e.g., Alex Edmans & Clifford G. Holderness, Blockholders: A Survey of Theory and Evidence,

European Corporate Governance Institute Finance Working Paper No. 475/2016 (Aug. 11, 2016) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2820976; Clifford G. Holderness, A Survey of Blockholders and Corporate Control, 9 Fed. Reserve Bank of N.Y. Econ. Pol’y Rev. 51 (Apr. 2003), https://www.newyorkfed.org/research/epr/03v09n1/0304hold.html.

16 Id. 17 See 15 U.S.C. § 78m(d); 17 C.F.R. §§ 240.13d-1, 240.13d-101. Conveniently as well for present

purposes, these same SEC rules require disclosure of an acquiring blockholder’s citizenship or place of organization. See 17 C.F.R. § 240.13d-101 (item #6).

18 A minor related problem involves the scenario of a 5% blockholder selling a small percent of shares and thus ceasing to be a 5% blockholder. SEC Rule 13d-2 requires an amendment for “any material increase or decrease in the percentage of the class beneficially owned.” 17 C.F.R. § 240.13d-2(a). Under this rule, an acquisition or disposition of one percent or more is per se deemed material, and an acquisition or disposition of beneficial ownership of less than one percent “may be material, depending upon the facts and circumstances.” Id. Moreover, the fact of a 5% blockholder disposing of sufficient shares to no longer meet the 5% threshold is potentially material on its own. Thus, generally, the best advice would be for the blockholder to file a termination amendment indicating the sell-down. In truth, many filers fail to do this, and issuers are aware of this problem. Consequently, when companies file their proxy statements for annual meetings, and must disclose “known” 5% holders under SEC Regulation S-K Item 403, see 17 C.F.R. § 229.403(a), they sometimes ask Schedule 13D filers whether they are still in fact 5% owners, and if not, do not include them in the proxy.

7

additional interest, we also calculated how many of these blockholders held ten percent, fifteen percent, or twenty percent of shares.19 We did not, however, examine scenarios involving multiple foreign blockholders having holdings below five percent each that, together, would add up to over five percent foreign ownership, except to the extent that SEC rules governing “groups” or “beneficial ownership” would trigger a filing under Section 13(d). For example, a corporation in which two distinct and unrelated foreign institutional holders each held four percent would not register as foreign-owned in our survey, despite these foreign entities controlling a combined eight percent. Again, our findings below thus represent a lower bound on the amount of foreign blockholder ownership and potential influence over large U.S. corporations.

C. Sample, data collection and analysis

We began with a list of the companies in the S&P 500 as of September 15, 2016.20 The S&P 500 is composed of publicly traded companies that are representative of the U.S. equity markets, “and through the markets, the U.S. economy.”21 While not every company in the S&P 500 is U.S.-based, more than 95% are, and they all have major U.S. operations. The S&P 500 companies are all “large cap” — companies that have a large market capitalization and correspondingly large role in the U.S. economy. At any one time, the S&P 500 has 500 constituent companies, but may contain more than 500 securities because some companies have more than one class traded in the U.S. equity markets. We then relied on publicly available data to determine each S&P 500 corporation’s top institutional holders and whether each holder met the above definition of a foreign investor. Specifically, for each equity security, we identified the top institutional holders as listed by Yahoo Finance (or MSN Finance, for those companies for which Yahoo lacked data) and noted those that held blocks of over five percent.22 We then determined whether those blockholders either were themselves foreign or were majority owned or controlled by a foreign entity. This investigation relied on a variety of public sources: the institutional holders’ own websites, Wikipedia entries, and other readily available web sources. We recorded the size (in percent of shares

19 These are obviously subsets — i.e., a 20% blockholder also counts as a 15% blockholder, a 10% blockholder, and a 5% blockholder.

20 For historical data on the composition of the S&P 500, see http://siblisresearch.com/data/historical-components-sp-500/ (last visited October 17, 2016).

21 S&P Global, The S&P U.S. Indices, Methodology (Aug. 2016), available at http://us.spindices.com/documents/methodologies/methodology-sp-us-indices.pdf?force_download=true (last visited October 17, 2016) at 1.

22 These sites ultimately draw their data from SEC filings as made available through the EDGAR database. We did not manually cross-check the data against SEC filings.

8

outstanding) of each foreign blockholder so identified. Ten percent of the data were randomly chosen and spot-checked by a second author for consistency and replicability. We have focused on institutional owners in this analysis. We do not count or attempt to classify as foreign or not individuals who own blocks (e.g., Jeff Bezos at Amazon) or trusts or closely held companies that are commonly used to hold blocks by individuals or families (e.g., the Walton family at Wal-Mart). As a result, as with our definition of “foreign investor” and our focus solely on blocksholders, our findings should be understood as putting a lower bound on the potential for foreign investor influence over U.S. public companies.

II. Findings The data reveal several interesting facts. As a broad top line result, Table 1 reports foreign institutional blockholder investors at different ownership thresholds:

Table 1. Foreign blockholder ownership

Foreign ownership

threshold Number of securities % of S&P 500

5% 47 9.3% 10% 9 1.8% 15% 5 1.0% 20% 3 0.6%

The numbers in Table 1 refer to single blockholders. In other words, the three corporations listed at the 20% threshold all have a single foreign blockholder with 20% or more of shares, and down the thresholds. The three companies with 20+% foreign blockholders are:

• General Growth Properties Inc. (NYSE: GGP), of which Canada-based Brookfield Asset Management owns 33.91%;

• Morgan Stanley (NYSE: MS), of which Japan-based Mitsubishi UFJ Financial Group, Inc. owns 22.3%; and

• Starwood Hotels and Resorts (NYSE: HOT), of which UK-based Omni

Partners LLP owns 21.1%. The data also show that three S&P 500 equities have multiple foreign blockholders, aggregating to more than ten percent in each case. These companies are:

9

• Level 3 Communications (NYSE: LVLT), of which Singapore-based Temasek Holdings (Private) Limited owns 18.17% and Singapore-based Singapore Technologies Telemedia owns another 18.17%;

• NASDAQ OMX Group (NASDAQ: NDAQ), which owns and operates the NASDAQ exchange and eight European stock exchanges, of which Swedish investment firm Investor AB owns 11.79% and Massachusetts Financial Services Co. (a wholly-owned subsidiary of Canadian insurance company Sun Life Financial) owns 9.02%; and

• Varian Medical Systems (NYSE: VAR), of which Loomis Sayles &

Company (a wholly-owned subsidiary of French asset management company Natixis Global Asset Management) owns 6.2% and Veritas Asset Management (UK) Ltd. Owns 5.53%.

By overall count, foreign institutions make up more than 30% of the institutions holding 5+% blocks in the S&P 500. In Table 2, we report the top ten most commonly found U.S. institutions with five percent or greater blocks in the S&P 500, and the top ten most commonly found foreign institutions with such blocks. As can be seen, U.S. institutions much more frequently have large ownership blocks, but one foreign institution (MFS) would appear in the top ten institutions overall, by block frequency. Even among U.S. institutions, only the top 28 hold more than two 5+% blocks — block ownership is itself highly concentrated.

Table 2. Ten Most Frequent U.S. and Foreign Block Holders

Top U.S. Institutions Top Foreign Institutions

Institution Number of 5+% Blocks in S&P 500 Institution Number of 5+%

Blocks in S&P 500

Vanguard 462 MFS 15 State Street 98 Baillie Gifford 3

Fidelity 91 Barrow, Hanley 3 T. Rowe Price 67 Macquarie Group 3 Capital Group 80 Daiwa Securities 3

Wellington 30 Harris Assoc. 2 JP Morgan 28 Loomis Sayles 2 Primecap 16 Norges Bank 2

Dodge & Cox 12 Standard Life 2 BNY Mellon 10 Brookfield 1

10

In Table 3, we report summary statistics on foreign institutional block ownership, conditional on the presence of any foreign block ownership. By construction, the smallest block (5.04%) is just over the SEC Section 13(d) threshold. The largest amount of foreign ownership — 36% — is at Level Three Communications, which as noted above has two Singapore blockholders. Conditional on the presence of one foreign blockholders, the median amount of foreign block ownership is 6%, and the average is roughly 9%. Foreign blocks represent 4.4% of all 5+% blocks in the S&P 500.

Table 3. Summary Statistics on Foreign Institutional Blocks

Minimum (by construction) 5.04% Maximum 36.34%

Median 6.30% Average 8.91%

Foreign 5+% blocks / all 5+% blocks 4.38% These statistics are conditional on the presence of any foreign institutional blockholding, i.e., for companies with at least one foreign institution owning 5+% of the company’s stock. In sum, foreign institutional blocks of 5+% or more are material, and common, type of ownership in the S&P 500. Most of the companies in that index do not have foreign institutional blockholders, and most institutional blockholders of such companies are not foreign. However, foreign blockholders have a significant presence at one in eleven of the largest U.S. public companies, foreign blockholding institutions are among the most frequent blockholders, and foreign institutions represent a large fraction (almost a third) of all institutions holding at least one block in the S&P 500.

III. Limitations, Implications and Future Research This is the first recent empirical analysis of the level of foreign institutional blockholder ownership of publicly traded corporations. The analysis has limitations. Because companies in the S&P 500 have large market capitalizations, the data may not be reflective of publicly traded corporations more broadly. On the one hand, blocks of larger companies by definition are larger in absolute value, which may constrain the ability of specific foreign investors from acquiring large blocks of stock of such companies. On the other hand, such companies are more prominent and may attract more foreign investment than other U.S. companies. Membership in the S&P 500 alone attracts institutional investment flows.23 It is thus uncertain whether other U.S. public companies have more, or less, foreign blockholder ownership.

23 E.g., L. Harris and E. Gurel, Price and Volume Effects Associated with Changes in the S&P 500 List: New Evidence for the Existence of Price Pressures, 41 J. Fin. 815-29 (Sep. 1986).

11

The data are almost certainly not reflective of non-publicly traded corporations. Such companies vary enormously in size and ownership, and there are many U.S.-based subsidiaries that are 100% owned by foreign investors, just as there are many U.S.-based companies with zero foreign ownership. Aggregate data from the Internal Revenue Service include over 83,000 tax returns filed by U.S. corporations that are controlled (i.e., have more than 50% of their stock) owned by foreign owners, with more than $12 trillion in assets in 2012.24 Prominent examples of such companies are not hard to find: Imbev, the parent company of Anheuser-Busch, the maker of a beer that is currently but temporarily being branded as “America,” is a Belgium-incorporated company listed on Euronext Brussels.25 U.S. subsidiaries of such companies are not included in our data because the U.S. subsidiaries are not U.S. public companies, and are not eligible for inclusion in the S&P 500, even if they are as large as companies in that index. Finally, our data capture a snapshot in time. Foreign blockholder investment almost certainly changes over time. In all likelihood, foreign block ownership has been rising in recent years. Such a time trend would be consistent with aggregate data from the Federal Reserve that show that portfolio investment by foreign investors grew from about 5% of all U.S. corporate equity (public and private) in 1982 to more than 20% in 2015.26 Despite these limitations, the data in this study are consistent with two general conclusions: (1) a material share of large publicly traded U.S. corporations – one in eleven – have at least one foreign institutional investor with 5+% of voting shares; and (2) higher foreign-blockholder investment levels (e.g., foreign blockholders with 10% or more), or combinations of multiple foreign blockholders, are neither common nor absent among such companies. The presence of large or multiple foreign blockholders makes plausible that some degree of foreign influence in U.S. elections (or in industries for which foreign ownership may be viewed as problematic from a policy perspective) may occur even among U.S. companies with apparently dispersed ownership. Future research could productively consider multiple related questions. How general are the findings reported here among smaller public companies? When if ever can block ownership be observed to provide foreign investors with a degree of influence over U.S portfolio companies? Are there any evident determinants of foreign block ownership, such as industry, size, or liquidity? Do companies with foreign blockholders have different corporate governance or

24 See https://www.irs.gov/pub/irs-soi/soi-a-coit-id1511.pdf (last visited October 17, 2016). 25 http://www.ab-inbev.com/investors/corporate-governance.html (last visited October 17, 2016). 26 Steve M. Rosenthal and Lydia S. Austin, The Dwindling Taxable Share of U.S. Corporate Stock, Tax

Notes (May 16, 2016) 923, 929 (Figure 3).

12

executive compensation characteristics? What are the trends in foreign ownership – are they increasing, as Federal Reserve data on foreign portfollio investment generally suggest? What if any implications would an increasing prevalence of foreign blockholders have for various legal and regulatory policies? Are foreign blockholders pursuing the same general agendas as U.S. institutional blockholders, or do they pursue distinctive patterns of policies in corporate governance and other aspects of ownership? At a minimum, we believe that our findings suggest the need for greater attention to the possible need for additional disclosures by U.S. public companies with large and potentially influential levels of foreign ownership. To be effective, these disclosures may need to be tailored to different regulatory settings. For example, in the context of political expenditures, it may not be sufficient to rely on existing SEC rules requiring disclosure of block ownership, since the links between those disclosures and the audience for point-in-time political communications are unlikely to be made by the relevant audiences in a reasonable time frame. For example, SEC filings can be delayed for up to ten days after a blockholder acquires five percent ownership, and in any event data on foreign ownership from such filings is unlikely to be salient to viewers of such communications.

* * *